|
Author
|
Topic: Learning may be Irreducibly complex
|
Nel
Member
Member # 614
|
posted 07. January 2003 18:05
Another molecular machine may need to be added to the list of IC systems that cause us to infer intelligent design. A recent discovery may uncover the "nano device" behind memory and learning.
Recent research shows that multiprotein structures associated with neurotransmitter receptors and cell-adhesion proteins are responsible for information storage during learning. As workers in the field began to study the signaling mechanisms responsible for changes in the "synaptic gap" , the complexity of those mechanisms grew into what they are now calling a molecular machine. The machine reads the neural code and initiates long-term changes in synaptic structure and function.
This new research shows that textbook illustrations of NMDAR signaling at the synapse was extremely oversimplified.
Professor Seth Grant stated:
quote:
Instead of floating in a sea of cytoplasm within the postsynaptic dendritic spine...many of the downstream signaling molecules involved in
NMDAR signaling are actually physically coupled to
NMDAR through a variety of protein–protein interactions and adaptor proteins.
NMDAR is part of a large multiprotein signaling machine.
The use of biochemistry is extremely important in confirming predictions concerning the complex, however, this has largely met with unsuccessful attempts to isolate the complex. However, recently, Husi et al. have shown the size to be 2000 kDa (2 MDa). An extremely large complex indeed.
quote:
Indeed, the high complexity of the complexes was evident by gel staining and subsequent proteomic analysis of specific proteins. Husi and Grant combined a mass spectrometry analysis of protein bands and immunoblotting with hundreds of antibodies to document the identity of about 75 proteins.
Among these 75 proteins they found a large amount of signal transduction proteins, which were noted to be required for the induction of Long Term Potentiation (LTP) and Long Term Depression (LTD). These could form signaling modules within the complex which make up what modern engineers would call a signal integration device commonly used in applications that produce a stream of pulsed signals.
Professor Seth Grant:
quote:
"This molecular machine, which we call the hebbosome, is found at the synapses of the brain, and seems to be important for learning. It converts electrical activity in nerves into changes inside the cells. It's made up of about 50 to 100 proteins, all bound together ... like a mini-computer."
This machine may perform computations similar to transcription factors, which, for example, position the RNA polymerase correctly at the promoter, aid in pulling apart the two strands of DNA to allow transcription to begin, and release RNA polymerase from the promoter into the elongation mode once transcription has begun.
More clues to it's IC nature may be in the following statement by Dr. Grant:
quote:
"But mutations in the genes that generated the proteins could make the machine malfunction, leading to profound learning disabilities in
children and animals."
and
quote:
Defects in Hebbosomes underpin some human cognitive impairments including learning disability.
The hebbosome is named after Donald Hebb. Hebb proposed that the underlying principle of associative learning is that events in the "outside" world that cause two neurons to be active at the same time favor the strengthing of synapses between them.
The proteins are broken up into 5 classes, glutamate receptors ,cell adhesion proteins, adaptors , signaling proteins and pathways, and
cytoskeletal proteins.
The definition of IC is as follows:
quote:
A system performing a given basic function is irreducibly complex if it includes a set of well-matched, mutually interacting, nonarbitrarily individuated parts such that each part in the set is indispensable to maintaining the system's basic, and therefore original, function. The set of these indispensable parts is known as the irreducible core of the system.
So by the definition of IC, if it turns out that there is a minimal set of core parts that contribute to the function, which in this case is the induction of synaptic activity, then it certainly is IC. Mutations in all 5 classes already distrupts this function, which is a good clue as to it's ICness.
Migaud M et. al. Enhanced long term potentiation and impaired learning in mice with mutant postsynaptic density-95 protein. Nature 396 (1998) 433-439
Husi H et. al. Proteomic analysis of NMDA receptor-adhesion protein signaling complexes. Nat Neurosci 3 (2000) 661-669
Nelson Alonso
[ 07. January 2003, 18:12: Message edited by: nanosoliton ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 08. January 2003 11:26
Nelson,
Some interesting findings. I suspect you are correct in concluding that these processes are IC with respect to Darwinian and neo-Darwinian processes. I would guess or speculate that these findings have two other important implications for evolutionary and biological design theory.
DEMONSTRATING NON-DARWINIAN EVOLUTION
One interpretation or implication of IC is design by an external designer. A second interpretation of IC is design by a non-Darwinian process. Learning, and the complex mechanisms underlying learning, provide, IMO, an excellent opportunity to study non-Darwinian evolutionary processes. This opinion is based on two observations.
First, we know that complex intercellular communications mechanisms must play an important role in all multi-cellular organisms. This suggests that it should be possible to identify the precursors or starting points for the evolution of learning mechanisms.
Second, it is possible to define with a great deal of precision and specificity the goal, purpose, or function of the learning mechanisms. Based on work I did years ago, it can be shown that learning mechanisms change or reprogram the algorithm controlling a neurons input-output processing. The more refined or complex the mechanisms, the more different algorithms that are possible. Using these techniques and concepts, it is possible to tie small changes in learning mechanisms to small changes in functionality.
It seems almost certain that the definable changes in purpose or functionality would not translate directly into changes in the survivability of the organisms. This, IMO, would suggest the operation of internal or secondary selection processes( selection other than Darwinian natural selection).
Again this is just speculation on my part. The evidence seems to suggest that the evolution of neurons and nervous systems occurred very rapidly and had a dramatic impact on life forms(the Cambrian explosion?) . The mechanisms underlying neurons and nervous systems would thus appear to be good candidates for demonstrating both ID and non-Darwinian evolutionary mechanisms.
WITHIN LIFE TIME BIOLOGICAL DESIGN
As discussed on another thread, developmental processes suggest the existence of within life-time design processes. The study of learning mechanisms, I understand, suggest that learning is a very complex developmental process. Learning involves literally millions of ‘turn gene on or off’ operations for each neuron over the life time of an organism. If development requires the presence of dynamic and teleological within lifetime design processes then clearly nervous system information processing and behavior involves dynamic and teleological within lifetime design processes. Given the mechanisms involved, the suggestion that behavior is inherited or front loaded would seem to be very difficult to justify.
An interesting extrapolation of this analysis would be the suggestion/prediction that human behavior and human nature are highly dynamic or highly modifiable.
The mechanisms underlying learning are, IMO, very interesting with a number of very interesting implications. An excellent topic to introduce for discussion.
IP: Logged
|
|
|
Printer-friendly view of this topic